3876
Vol. 53
H. M. HUFFMAN, G. S. PARKS AND MARK BARMORE
[CONTRIBUTION FROM THE DEPARTMENT O F CHEMISTRY OF
STANFORD
UNIVERSITY ]
THERMAL DATA ON ORGANIC COMPOUNDS.
X. FURTHER STUDIES ON THE HEAT CAPACITIES, ENTROPIES AND FREE ENERGIES OF HYDROCARBONS* BY
HUGH M. HUFFMAN,* GEORGE s. PARKS3 AND MARKBARMORE' RECEIVED JULY 3, 1931
PUBLISHED OCTOBER5, 1931
I n four earlier papers5 heat capacity data have been presented for about forty hydrocarbons and, in so far as possible, the corresponding entropies and free energies have been calculated. Certain pronounced regularities in these data have also been noted. I n the present paper, which brings our studies on hydrocarbons to an end. we shall present similar thermal data for the following twenty compounds: propylene, n-butane, n-hexane, n-octane, n-nonane, n-decane, n-undecane, n-dodecane, methylcyclopentane, 1,2-dimethylcyclopentane,pseudocumene, durene, isodurene, prehnitene, $-cymene, n-butylbenzene, pentamethylbenzene, p-methylnaphthalene, anthracene and phenanthrene. Four of these (n-hexane, noctane, n-nonane and n-decane) were studied to some extent in the first paper referred to above but our new heat capacity data, obtained with purer hydrocarbon samples, are much more nearly complete and reliable. In general, our new entropy and free energy results confirm in a very satisfactory manner the conclusions reached in the earlier papers.
Materials Propylene.-This hydrocarbon was very carefully prepared for us by Dr. Gerald van de Griendt of the Shell Development Company. Extremely pure isopropyl alcohol was dehydrated by phosphoric acid and the resulting propylene, when purified, was estimated to contain less than 0.1% impurities. The sample had an extremely sharp melting point a t 88.2"K. n-Butane.-A large quantity of commercial n-butane (impurities guaranteed less than 1%) was obtained from the Carbide and Carbon Chemical Corporation. This material after three fractional distillations in a special still gave the product which was used in our measurements. n-Hexane, %-Octane, n-Nonane, %-Decane, n-Undecane and n-Dodecane.-These samples constituted part of the materials prepared by Shepard, Henne and Midgley, 1 This paper contains results obtained in an investigation of the heat capacities and free energies of some typical hydrocarbon compounds, listed as Project No. 29 of the American Petroleum Institute Research. Financial assistance in this work has been received from a Research fund of the American Petroleum Institute donated by The Universal Oil Products Company. This fund is being administered by the Institute with the cooperation of the Central Petroleum Committee of the National Research Council. 2 American Petroleum Institute Research Associate. Director, Project No. 29. 4 American Petroleum Institute Research Assistant (part-time). Parks, Huffman and Thomas, TRISJOURNAL, 5 2 , 1032 (1930); Huffman, Parks and Daniels, ibid., 52, 1547 (1930); Huffman, Parks and Thomas, ibid., 5 2 , 3241 (1930); Parks and Huffman, ibid., 52, 4381 (1930).
Oct., 1931
THERMAL DATA ON ORGANIC COMPOUNDS.
X
3877
and for a full description of their preparation and properties the reader is referred to the recent paper8 by these investigators. It is sufficient for us to say here that all of these samples showed very sharp melting points and were evidently of extreme purity. We likewise made measurements upon the samples of n-pentane and n-heptane prepared in the same series. However, as our earlier measurements upon these two compounds were also made with very pure materials and check the new values to better than 1%, we have omitted these data for n-pentane and n-heptane in preparing the present paper. Methylcyclopentane and 1,2-Dimethylcyclopentane.-These two hydrocarbons were kindly loaned to us for our heat-capacity measurements by Professor G. Chavanne, in whose laboratory a t the University of Brussels they had been prepared. The methylcyclopentane boiled a t 71.5-71.7' (at 752.5 mm.) and had a very sharp melting point a t - 143.0', The 1,2-dimethylcyclopentane (supposedly the trans form) boiled a t 91.8"; it melted a t -119.0" with considerable premelting. Pseudocumene, Durene, Isodurene, Prehnitene and Pentamethylbenzene.-The samples of these five hydrocarbons were kindly loaned to us by Professor Lee Irvin Smith of the University of Minnesota. Details concerning them will not be given here, as their preparation and properties have been fully described elsewhere.' It is sufficient for us to add that all of these materials were of satisfactory purity for. our measurements. n-Butylbenzene and p-Cymene.-High grade Eastman materials were subjected to several fractional distillations in a special still. The final products had narrow boiling ranges: 12-butylbenzene, 183.1-183.5'; p-cymene, 176.7-177.0". pMethylnaphtha1ene.-This was a highly purified German material which was subjected to eight fractional crystallizations in our own laboratory. The melting point of the final product was 34.1". Anthracene.-The anthracene was obtained from Eastman's material (m. p. 213 ") by six fractional crystallizations from benzene. The final product was very slightly yellow and had a melting point of 215.0". Further attempts to purify the material were unsuccessful. Phenanthrene.-Kahlbaum's phenanthrene was subjected to fourteen crystallizations from ethyl alcohol. Our final sample melted a t 97.7'.
Experimental Results In principle, the method of Nernst was employed with an aneroid calorimeter in determining the "true" specific heats and the fusion data. The apparatus and details of experimental procedure have been fully described in other places.* In view of the accuracy of the various measurements involved, the error in the experimental values thereby obtained is probably less than lye, except in so far as impurities in a hydrocarbon sample may cause premelting or otherwise influence the results. As liquid propylene has a vapor pressure of 7 or 8 atm. at room temperature, the filling of our thin-walled calorimeter can and its subsequent installation in our apparatus in the usual fashion was hardly feasible in this case. Accordingly we first installed an empty calorimeter can, which Shepard, Henne and Midgley, THISJOURNAL, 53, 1948 (1931). Smith and Lux, ibid.,51, 2994 (1929); Smith and MacDougall, ibid., 51, 3001 (1929); Smith and Lund, ibid., 52, 4144 (1930). Parks. ibid. 47. 338 (1925); also Parks and Kelley, J . Phys. Chem., 30, 47 (1926).
3878
VOl. 53
H. M. HUFFMAN, G. S. PARKS AND MARK BARMORE
was equipped with a German-silver entry tube, about 1.0 mm. inside diameter and 100 cm. long. This entry tube ran outside of the calorimeter system and was connected through a valve system to a monel-metal bomb of about 40-cc. capacity. The bomb was equipped with a vacuumand pressure-tight valve so that it could be disconnected and weighed with its contents. By connecting the bomb filled with propylene to the entry tube and opening the valve, any desired quantity of propylene could be distilled into the calorimeter can and condensed by the maintenance of liquid-air temperatures around the calorimeter system. For emptying the calorimeter after a series of heat capacity determinations, this distillation process was easily reversed. Thus the weight of hydrocarbon used could be determined before, and checked after, a series of measurements with an accuracy of 0.05%. The specific heats and the fusion data, expressed in terms of the 15' calorie9 and with all weights reduced to a vacuum basis, appear in Tables I and 11. For comparison with these values the literature contains only TABLE I
SPECIFIC HEATS Temp., OK. Cp per g.
68.9 0.273
72.0 0.280
Temp., OK. CP Per g. Temp., OK. C p per g.
93.1 0.523 189.5 0.502
93.5 0.522 210.3 0.512
Temp., OK. C p per g.
69.1 0.212
74.9 0.226
Temp., "K. C, per g.
114.3 0.35
119.9 0.37
Temp., O K . CPper g.
139.7 0.467
150.2 0.469
Temp., OK. C,per g.
93.4 0.219
99.0 0,227
Temp., OK. Cpper g.
188.8 0.472
217.8 0,482
PROPYLENE: Crystals 76.5 73.5 76 7 0.286 0.296 0.295 Liquid 98.6 108.7 0.516 0.504
81.6 0.317
125.2 0.494
144.8 0.490
153.9 0.490
%-BUTANE:Crystals I 92.5 89.5 83.0 0.238 0.252 0.256
95.6 0.264
97.0 0.267
187.0 0.484
190.1 0.483
230.0 0.506
HEXANE: Crystals 115.0 131.4 145.5 0.251 0.274 0.295
154.8 0.309
163.5 0.328
165.3 0.492
Crystals I1
@
Liquid 152.5 170.2 0.472 0.474
Liquid 275.4 276.2 0.521 0.523
261.8 0.533
293.5 0.536
The factor 0.2390 has been used in converting the joule to the 15" calorie.
Oct., 1931
THERMAL DATA ON ORGANIC COMPOUNDS.
X
TABLE I (Continued) Crystals 140.2 170.3 0.263 0.300
%-OCTANE:
92.4 0.198
97.3 0.207
110.8 0.226
Temp., "K. Cp per g.
227.0 0.483
250.9 0.490
Liquid 275.0 286.6 0.508 0.517
Temp., OK. Cp per g.
92.8 0.195
97.6 0.203
113.8 0.227
136.8 0.256
228.3 0.489
233.7 0.489
Liquid 245.0 259.4 0.491 0.498
179.2 0.312
188.7 0.327
197.7 0.346
150.1 0.271
163.1 0.288
179.3 0.309
187.7 0.322
275.3 0.508
282.8 0.514
289.9 0.518
297.9 0.523
298.3 0.526
W N O N A N E :Crystals
%-DECANE: Crystals Temp., "R. Cpper g. Temp., OK. c, Per g.
90.8 0.188 0.196
102.2 0.204
107.0 0.210
113.9 0.219
120.7 0.228
130.1 0.239
139.5 0.250
150.0 0.262
159.6 0.273
170.1 0.285
180.2 0.297
190.6 0.310
200.2 0.323
210.5 0.338
220.6 0.355
Temp., OK. Cp per g.
251.2 0.495
262.1 0.500
Liquid 275.2 281.3 0.509 0.512
288.8 0.517
297.7 0.523
91.3
WUNDECANE:Crystals 190.1 0.310
208.4 0.338
136.4 0.239
148.0 0.252
234.0 0.362
243.5 0.380
92.0 0.188
97.5 0.1g6
112.4 0.215
140.2 0,250
158.1 0.271
Temp., OK. C P Per g.
258.5 0.503
274.9 0.511
Liquid 283.4 290.8 0.515 0.520
298.0 0.524
Temp., O K . Cpper g. Temp., "K. CP Per g.
93.3 0.184 171.0 0.277
n-DODECANE: Crystals 106.3 113.7 124.1 0.202 0.212 0.225 186.3 197.6 211.4 224.0 0.295 0.309 0.326 0.344
Temp., "K. CP Per g.
275.1 0.510
282.9 0.514
Temp., OK. CP Per g.
92.2 0.183
METHYLCYCLOPI.:NTANE: Crystals 97.0 103.3 110.6 117.4 123.3 0.190 0.198 0.208 0.218 0.230
Temp., OK. Cp per g.
139.0 0.352
169.5 0.357
99.5 0.194
159.3 0.264
Liquid 289.7 297.7 0.518 0.521
Liquid 189.2 210.3 0.365 0.376
230.0 0.388
251.3 0.404
275.1 0.424
293.7 0.447
3880
VOl. 53
H . M. HUFFMAN, G. S. PARKS AND MARK BARMORE
Temp., "K. C p per g.
TABLE I (Continued) 1&DIMETHYLCYCLOPENTANE: Crystals 99.4 107.0 114.8 122.3 129.3 93.4 0.191 0.200 0.211 0.221 0.232 0,263
Temp., "K. C p per g.
161.5 0.364
175.2 0.371
244.6 0.414
275.4 0,439
284.1 0.446
294.2 0.456
Temp., OK. CP per g.
93.7 0.158
PSEUDOCUMENE: Crystals 100.0 115.1 123.7 140.8 0.166 0.185 0.195 0.214
163.0 0.240
180.7 0.259
205.1 0.307
Temp., OK. Cp per g.
239.5 0.388
246.9 0.392
Liquid 260.5 277.0 0.400 0.412
277.4 0.411
283.6 0.415
297.3 0.422
Temp., OK. C p per g. Temp., OK. C, per g.
92.2 0.156
98.7 0.166
DURENE:Crystals 106.4 114.4 123.2 0.176 0.186 0.197
137.2 0.214
154.2 0.233
170.1 0.251
189.5 0.271
202.3 0.285
219.5 0.302
257.2 0.338
277.3 0.361
284.4 0.369
297.1 0.383
Temp., OK. C p per g.
92.4 0.162
103.3 0.176
ISODURENE: Crystals 121.7 140.4 160.2 0.199 0.222 0.245
189.3 0.275
200.3 0.287
210.7 0.302
Temp., OK. CP per g.
255.3 0.401
275.7 0.414
Temp., "K. C, per g.
91.0 0.149
95.5 0.154
100.1 0.160
177.5 0.252
208.3 0,296
223.6 0.328
Temp., OK. Cp per 6.
276.5 0.416
281.8 0.417
Liquid 286.5 291.9 0.418 0.420
Temp., OK. C, per g.
92.2 0.156
96.9 0.162
-CYMENE:Cry5 als 103.8 111.3 120.3 0.170 0.179 0.190
139.8 0.212
158.9 0.233
179.8 0.269
Temp., " X . C, per g.
210.8 0.367
215.9 0.370
Liquid 228.2 243.3 0.376 0.384
259.6 0.393
280.7 0.409
291.0 0.417
297.1 0.421
Temp., "K. C, per g.
94.0 0.154
n-BUTYLBENZENE: Crystals 99.6 105.6 115.3 127.5 0.160 0.166 0.177 0.192
139.2 0.205
151.3 0.218
161.0 0.231
Temp., O K . Cp per g.
191.9 0.369
195.8 0.371
Liquid 210.6 224.8 0.377 0.383
275.5 0.411
287.9 0.420
298.2 0.428
Liquid 195.0 210.0 0.385 0,393
242.1 0.324
Liquid 281.6 288.6 0.417 0.422
297.1 0.428
Crystals 122.4 145.8 0.185 0.212
PREIINITENE:
255.0 0.400
Oct., 1931
THERMAL DATA ON ORGANIC COMPOUNDS.
Temp., OK. Cpperg.
91.7 0.169
Temp., OK. CPper g.
189.4 0.261
Temp., OK. Cp per g.
303.6 0.447
Temp., "K. C p per g.
93.8 0.121
Temp., O K . CP Per g.
310.4 0.383
TABLE I (Concluded) PENTAMETHYLBENZENE: Crystals I 105.2 113.1 122.6 141.2 0.182 0.188 0.197 0.213 205.0 213.0 231.9 242.4 0.277 0.286 0.304 0.316
X
3881
149.9 0.222
159.2 0.230
174.8 0.245
258.4 0.334
265.0 0.341
283.8 0.405
225.2 0.238
253.3 0.268
272.4 0.299
Crystals I1
@-METHYLNAPHTHALENE: Crystals 107.1 0.133
135.5 0.157
157.1 0.175
196.6 0.210
Liquid
94.4 0.095
102.4 0.101
193.2 0.177
210.2 0.194
Temp., OK. CP Per g.
93.4 0.097
Temp., OK. CP Per g.
210.4 0.195
ANTHRACENE:Crystals 110.8 118.4 124.8 0.107
0.113
0.117
142.4 0.131
158.2 0.146
173.3 0.159
228.4 0.211
244.6 0.227
254.4 0.236
275.8 0.257
282.5 0.264
297.2 0.278
157.9 0.147
179.7 0.166
190.5 0.176
290.6 0.300
297.5 0.313
304.4 0.325
PHENANTHRENE: Crystals 100.1 116.7 127.5 137.9 0.102 0.114 0.122 0.130 232.0 252.2 277.4 283.9 0.216 0.236 0.265 0.277
very meager data. Mabery and Goldsteinlo have studied the heat capacities between 0 and 50" of a number of hydrocarbons, including nhexane, n-octane, n-nonane, n-decane, n-undecane and n-dodecane. They used an ice calorimeter and worked on materials derived from petroleum. Their result for n-octane at 25" is 4% below our curve; for the other compounds their values deviate by from 2.0 to 4.5% from our curves. Referring to our own previous work, we find that these new values for normal hydrocarbons are in good agreement with our earlier data as far as the liquid state is concerned. On the other hand, the new values for the n-hexane and n-octane crystals are for the most part considerably lower than the older ones, which were apparently affected by premelting to a greater extent than we at first realized. Conversely, our new fusion values for n-hexane and n-octane are 3.5 and 2.7% higher than the corresponding earlier data. The specific heat values for liquid propylene show rather abnormal behavior. There is a 7% decrease as we proceed up the temperature scale from 93 to 125'K., whereas for most liquids there is a small but definite lo Mabery and Goldstein, Am. Chem. J., 28, 69 (1902).
3882
€ M. I . HUFFMAN, G.
S. PARKS AND MARK BARMORE
VOl. 53
TABLE 11 FUSION DATA^ Substance
M. p.,
OK.
Heat of fusion (cal. per 9.) 1st result 2d result Mean
Propylene 88.2 16.62 16.75 16.67 ... 17.98 n-Butane 134.1 17.98 ... 36.14 n-Hexane 177.9 36.14 ... 43.21 %-Octane 215.8 43.21 n-Nonane 219.2 41.24 41.19 41. 22b n-Decane 243.1 48.34 48.34 48.34 n-Undecane 247.2 34.24 34.00 34.12 ... 51.33 n-Dodecane 263.5 51.33 Methylcyclopentane 130.1 19.52 19.58 19.55 1,2-Dimethylcyclopentane 154.1 15.66 ... 15.66 ... 25.15 Pseudocumene 228.6 25.15 Isodurene 248.6 23.06 23.02 23.04* Prehnitene 265.4 20.01 19.99 20.00 $-Cymene 204.2 17.21 17.20 17.20 n-Butylbenzene 184.6 19.55 19.55 19.55 8-Methylnaphthalene 307.2 20.09 20.12 20.11 I n the calculation of these fusion values, the marked rise in the specific heat of the crystals as the melting point is approached was attributed to premelting; and the heat absorbed in this region in excess of that obtained by extrapolation of the specific heat data at lower temperatures was added to the heat absorbed a t the melting point. This value includes the heat effect for a solid transition taking place a few degrees below the melting point. TABLE I11 TRANSITION DATA Substance
n-Butane n-Undecane Pentamethylbenzene
Transition point, OK.
107.0 236.1 296.8
Heat of transition (cal. per 6.1 1st result 2d result Mean
8.7 9.67 3.19
.. 9.71
..
8.7 9.69 3.19
increase in heat capacity with rising temperatures. For this reason we a t first feared that the use of our special calorimeter can with its long entry tube was affecting our results in some unaccountable manner. However, a second series of determinations made on n-butane with the special calorimeter yielded values which were in excellent agreement with earlier data obtained in the usual way. Thus the propylene behavior is apparently real. Five of the hydrocarbons, n-butane, n-nonane, n-undecane, isodurene and pentamethylbenzene exhibited definite transitions in the solid state with quite appreciable heat effects. In three instances these heats of transition were directly measured, and the data thus obtained are recorded in Table 111. In the case of n-nonane the transition came only 2.5" beloq the melting point. The situation is illustrated graphically in Fig. 1, where we have plotted the time-temperature curve for the transition and fusion processes. It is noticeable that the horizontal steps for the
Oct., 1931
THERMAL DATA ON ORGANIC COMPOUNDS.
X
3883
two processes stand roughly in the ratio of 2 to 5. However, i t was hardly practical to determine accurately the heats of transition and fusion apart from one another. Therefore] the two quantities are lumped together in the fusion value given in Table 11. For most practical purposes, such as entropy calculations, there is no real objection to such a procedure. We have acted similarly in the case of the isodurene transition, which came about 10” below the melting point. In this case the ratio between the transition heat and the fusion heat was about 2 to 9, and our heat of fusion is therefore about 2500 cal. per mole. This result is in satisfactory agreement with the estimate (2550 cal. per mole) obtained by Smith and MacDougall from freezing point measurements on the durene-isodurene system.’
0 Fig. 1.-The
40 60 80 100 Time in minutes. time-temperature curve for the transition and fusion of n-nonane. 20
Discussion Entropies of the Compounds.-Using the data contained in Tables I, I1 and I11 in conjunction with the third law of thermodynamics] we have calculated the molal entropies a t 298.1OK. for the various hydrocarbons. In these calculations we have employed the extrapolation method of Kelley, Parks and Huffmanl’ for estimating the entropy increase for the crystals, Col. 2 of Table IV, from 0 to 90’K. The various entropy increments from 90 to 298.1 OK., which appear in Cols. 3, 4 and 5 of the table, were obtained by the usual methods directly from the experimental data. The results for the total entropy in calories per degree appear under the heading ‘S298 experimental” in the sixth column. I1 Kelley, Parks and Huffman, J. Phys. Chem., 33, 1802 (1929).
3884
H. M. HUFFMAN, G. S. PARKS AND MARK BARMORE
Vol. 53
Two of these hydrocarbons, propylene and n-butane, are gases at 298°K. and 1 atm. pressure. Accordingly, in compiling the table, we have estimated their entropies of vaporization at the respective boiling points (223.3 and 273.7'K.) by use of the corresponding vapor pressure data of Burrell and Robertson.I2 The entropy increases in the gaseous state from the boiling points to 298.1'K. were then estimated on the assumption of C, = 14.0 cal. per mole for propylene and C, = 22.3 cal. per mole for nbutane. While these assumptions are rather arbitrary guesses, they cannot involve us in serious errors, as the magnitude of these increments is small. In the cases of durene and P-methylnaphthalene, compounds which are normally crystalline a t 298'K., i t has seemed desirable for purposes of TABLE IV ENTROPIES OF THE HYDROCARBONS PER MOLE -CrystalsSubstance
O-SO'R.
Above
90'K.
Fusion
Liquid
S&. experi-
Szas
mental predicted
...
12.0" * . . 7.96 19.26 63.1b Propylene (gas) 11.7 12.3lC 7.79 23.13 54.9 55.8 %-Butane (liquid) 20.36 74.4d . . , . . . ... n-Butane (gas) 15.33 15.88 17.50 21.90 70.6 71.2 %-Hexane 18.14 26.68 22.85 18.33 86.0 86.6 %-Octane 19.86 30.23 24.10 19.69 93.9 94.3 n-Nonane 22.00 37.56 28.27 14.65 102.5 102.0 %-Decane 23.92 50.51 21.55 14.91 110.9 109.7 n-Undecane 25.10 49.10 33.16 10.78 118.1 117.4 n-Dodecane 13.83 6.36 12.64 26.38 59.2 Methylcyclopentane 16.67 11.98 9.97 25.86 64.5 . . . 1,2-Dimethylcyclopentane 16.68 25.02 13.21 12.80 67.7 67.6 Pseudocumene ... 58.7 18.33 40.37 ... Durene (solid) ... ... 13.13 ... 71.8 75.3 Durene (liquid) 19.25 32.35 12.43 10.03 74.1 75.3 Isodurene 18.24 34.59 10.11 6.51 69.5 75.3 Prehnitene 19.12 23.16 11.30 19.70 73.3 75.3 p-Cymene 18.87 18.60 14.20 25.10 76.8 75.3 n-Butylbenzene 70.3 . . . 24.59 45.63' Pentamethylbenzene (solid) . . . . . . 48.7 . . . 15.74 33.00 @-Methylnaphthalene(solid) 57.9 ... ... ... 9.2 ... @-Methylnaphthalene(liquid) ... 49.6 ... 14.98 34.60 ... Anthracene (solid) . . . 50.6 ... 15.58 35.05 Phenanthrene (solid) This value a The entropy of propylene crystals a t the melting point (88.20K.). includes 20.0 E. U. for the entropy of vaporization of propylene at the normal boiling and 3.9 E.U. for the entropy increase of the gas between this tempoint (225.3"K.) perature and 298.1"K. This value includes the entropy increase for both crystalline This forms as well as the entropy effect (4.72E. U.) for the transition between them. value includes the entropy of vaporization a t the boiling point (20.44E. U.)and 1.91 E. U. for the entropy increase of the gas between 273.7and 298.1OK. This value includes 1.85E. U. for an entropy of transition a t 297°K.
...
...
...
... ...
11
Burrell and Robertson, THISJOURNAL, 37, 2188 (1915).
...
Oct., 1931
THERMAL DATA ON ORGANIC COMPOUNDS.
X
3885
comparison to have values for the entropy as a liquid, even though the liquid state is unstable a t this temperature. Accordingly we have roughly calculated their entropies of fusion a t 298” by the method previously employed by Huffman, Parks and DanielsS6 For this purpose we have used the heat of fusion of durene recently obtained by Smith and MacDougall’ and our own result for 0-methylnaphthalene.
n. Fig. 2.-The molal entropies of some normal paraffins plotted against the number of carbon atoms in the molecule.
In the preceding papers6 it has been shown that the entropy of a liquid paraffin or benzenoid hydrocarbon can be calculated quite accurately by the empirical equation, s298 = 25.0 7.7n - 4 . 5 ~ 19.5p. Here n and p represent, respectively, the number of aliphatic carbon atoms and phenyl groups in the molecule, and r ordinarily refers to the number of methyl branches attached on the main aliphatic chain. I n the present study we have applied this equation, wherever possible, to obtain the values of “ S 2 9 8 predicted” in the last column of the table. I n the case of the normal paraffins these are in excellent agreement with the experimental results, as a comparison of the tabulated data will show. This agreement is also shown graphically in Fig. 2, where the circles represent the experimental values of s298 for the various normal paraffins studied in this and in the preceding papers and the straight line represents the graph of our empirical entropy equation. I n the case of the benzenoid compounds the agreement between the experimental results and “predicted &98” values is, in general, considerably poorer, although even here the
+
+
3886
Vol. 53
H. M. HUFFMAN, G. S. PARKS AND MARK BARMORE
empirical equation serves to give a rough approximation of the entropy of any hydrocarbon. The Free Energies of Fourteen Hydrocarbons.-We have also calculated the free energies of fourteen of these hydrocarbons by means of the third law of thermodynamics and the fundamental equation, AF = A€€- T A S . The essential data are given in Table V. For obtaining the AH of formation of propylene we have used the heat of combustion reported in TABLE V THERMAL DATAAT 298.1 “K. The 15” cal. is used throughout and all weights have been reduced to a vacuum basis. Substance
Heat of combustion a t constant pressure
Propylene (g) %Butane ( 9 ) %-Hexane n-Octane %-Decane M ethylcyclopentane
1,2-Dimethylcyclopentane Pseudocumene Durene (s) $-Cymene n-Butylbenzene Pentamethylbenzene Anthracene (s) Phenanthrene (s)
492,200 686,400 990,200 1,304,500 1,615,200 937,500 1,090,200 1,239,800 1,393,000 1,401,800 1,401,900 1,553,400 1,693,100 1,692,200
AHm,
Cal.
$ 4,550
-32,110 -53,410 -64,210 -78,610 -37,800 -47,650 -18,220 -27,570 -18,770 -18,670 -29,720 +32,200 $31,300
ASm, E. U.
- 34.5 - 87.0 -155.6 -205.5 -264.0 -136.0 -163.2 -131.4 172.9 -158.3 -154.8 193.8 -124.7 - 123.7
-
-
AF&
Cal.
+14,800 - 6,200 - 7,000 - 3,000 - 2,900 2,700 1,000 +21,000 $24,000 +28,400 +27,500 $28,100 $69,400 $68,200
+ +
the early days by Thomsen,l3 and for anthracene and phenanthrene, Stohmann’s v a 1 ~ e s . l ~For pseudocume we have employed the result obtained by Richards and Barry and more recently revised by Swietoslawski and Bobinska;lS and for n-butylbenzene we have adopted their corresponding result for n-propylbenzene increased by 156,300 cal. (the observed increase in the heat of combustion per CH2 increment). In the absence of any reliable combustion values for n-butane and n-decane, we have substituted Thomsen’s result for isobutane and that of Richards and for diisoamyl. Such a substitution seems entirely justifiable because the U. S. Bureau of Standards has recently shown that the nine isomeric heptanes have practically identical heats of combustion and that n-octane 13
Thomsen (translation by Burke), “Thermochemistry,” Longmans, Green and
Co., London, 1908, p. 441. 14 Landolt-Bbrnstein-Roth-Scheel, “Tabellen,” Julius Springer, Berlin, 1923, p. 1590. 15 Swietoslawski and Bobinska, THIS JOURNAL, 49, 2478 (1927). The values have been increased by 0.0117,, as suggested by Verkade and Coops, Rec. trav. chim., 46, 910 (1927).
Oct., 1931
THERMAL DATA ON ORGANIC COMPOUNDS.
X
3887
and a branched isomer, 2,2,4-trirnethylpentane,differ in this particular by less than O.la/O.l6 In the cases of the remaining seven compounds we have taken the heats of combustion as given in the “International Critical Tables.”17 For our present purposes we have converted all these combustion data to 298.1’K. The A H 2 9 8 values were then calculated by use of 68,31018and 94,24019 cal. for the heats of combustion of hydrogen and graphitic carbon, respectively. Column 4 contains the entropy of formation of each compound, which is simply the difference between its SZg8and the corresponding values for the entropies of the elements contained therein. For this purpose the respective entropies of carbon and hydrogen were taken as 1.320and 15.6221E. U. per gram atom. The molal free energies appear in the last column of the table. For the most part the accuracy of these values is largely limited by the accuracy of the combustion data employed. I n the case of n-octane, where the heat of combustion has been determined recently by the U. S. Bureau of Standards, the free energy value is probably good to 1000 or 1500 cal. On the other hand, in the cases of durene, pentamethylbenzene, anthracene and phenanthrene, where the heats of combustion were determined by Stohmann and his co-workers in the early days, the combustion values may be in error by as much as four or five thousand calories; and therefore the free energies are uncertain to this extent. For the remaining compounds the values are probably reliable to within two or three thousand calories. On the whole these free energies are in excellent accord with the various conclusions presented in the preceding papers. However, in making comparisons with the previous data i t must be borne in mind that we are here employing different values for the heats of combustion of carbon and hydrogen and for the entropy of hydrogen. Our new values for the combustion of carbon and hydrogen are 30 and 20 calories, respectively, below the former ones, and serve to increase (algebraically) the corresponding free energy values by 50 calories per CH2 increment or, for example, by a total of only 420 calories in the case of n-octane. On the other hand, the alterations in free energy produced by the use of Giauque’s new entropy value for hydrogen are much more important, amounting t o an increase of 243 calories for each gram atom of hydrogen involved. This corresponds to 486 calories per CH2 increment, or 4370 calories in the case of n-octane. Fortunately the revision of our earlier data is very simple and may be easily made by any reader interested in a given free energy value. Kharasch, Bureau of Standards Journal Research, 2, 373 (1929). “International Critical Tables,” Vol. V, p. 163. I* Rossini, Bureau of Standards Journal Research, 6 , 34 (1931). Roth and Naeser, Z. Electrochem., 31, 461 (1925). We here have reduced their value to a vacuum basis. 2o Lewis and Gibson, TAISJOURNAL, 39, 2581 (1917). 21 Giauque, ibid., 52, 4816 (1930).
3888
EVAN W. MCCHESNEY AND CLEMMY 0. MILLER
VOl. 53
In the present study we have calculated a AF,& value for one or more representatives of each of the four important classes of hydrocarbons. The numerical results serve to illustrate that the order of decreasing thermodynamic stability (i. e., increasing free energy of formation) a t room temperature is: (1) paraffin, ( 2 ) naphthene, (3) olefin, (4)aromatic hydrocarbons. Before concluding, the authors wish to thank the Shell Development Company, Dr. Albert L. Henne, Professor G. Chavanne and Professor Lee Irvin Smith for the valuable hydrocarbons which made this research possible.
Summary 1. The specific heats of twenty hydrocarbons have been measured over a wide range of temperatures. Heats of fusion and of transition have also been determined in the case of seventeen of these compounds. 2 . The entropies of the twenty hydrocarbons have been calculated from these heat capacity data. In general the results are in good agreement with the corresponding values calculated by an empirical equation developed in preceding papers. 3. The corresponding free energies for fourteen of these hydrocarbons have also been calculated. The order of decreasing thermodynamic stability a t 298' K. is (1) paraffin, (2) naphthene, (3) olefin, (4) aromatic hydrocarbons. STANFORD UNIVERSITY,CALIFORNIA [CONTRIBUTION FROM
THE
CHEMISTRY LABORATORY OF NORTHWESTERN UNIVERSITY MEDICALSCHOOL]
STUDIES ON PROTEINS I N LIQUID AMMONIA.
I*y2
BY EVAN W. MCCHESNEY AND CLEMMY 0. MILLER RECEIVED JUNE 20, 1931
PUBLISHED OCTOBER 5, 1931
During the last few years there has been an increasing interest in the study of proteins in non-aqueous media. Granacher3 has studied the reactions of proteins and polypeptides in ethyl alcohol a t 170'; Fodor and Epstein4 have studied the decomposition of gelatin by acetic anhydride. 1 This article is taken from the dissertation presented by Evan W. McChesney to the Graduate School of Northwestern University in partial fulfilment of the requirements for the degree of Doctor of Philosophy. 2 This is an abstract of two papers, one of which was presented before Section C of the American Association for the Advancement of Science, Cleveland, Ohio, and the other before the Organic Division of the American Chemical Society in Indianapolis, Indiana, April, 19'31. 8 Granacher, Helv. Chim. Acta, 8 , 784 (1925). 4 Fodor and Epstein, Z. physiol. Chem., 171, 222 (1927): Biochem. Z., 200, 211 (1928); 228, 310 (1930).
